Research On Dynamic Interception Coalition Formation Mechanism And Method In Networked Air Defense Missile Systems

Air attack and anti-air warfare are indispensable to the future war. The development of modern air attack techniques and weapons put forward new challenges to air defense missile systems. The distributed flexible network-centric architecture represents the superior development of the air defense missile systems.Interception coalition (IC) formation is one of the cutting-edge problems in Networked Air Defense Missile Systems (NADMS) research, aiming to allocate the most suitable weapons and guidance radars to the targets, so as to maximize the effectiveness of NADMS. Based on the combination of centralized and distributed command, this dissertation focuses on addressing centralized and distributed interception coalition formation method. The research includes the following parts:(1) IC formation problem is described as an operation research and integer programming problem, a combinatorial optimization problem and an economics game problem. Hypotheses for research are also given based on the background of NADMS. After that, a leveled decision framework is designed, as well as the distributed decision method, behavior models and information interaction frameworks for the distributed command and control nodes. Some key techniques in the IC formation problem research are pointed out.(2) A constrained optimization model for centralized IC formation problem is designed. Aiming to solve the model, the particle swarm optimization (PSO) is chosen based on the standards of capability, quality and efficiency of a method. To solve the premature convergence problem, which PSO suffers from in complex optimization problems, PSO is analyzed from a new aspect, which is cognitive psychology. A creative thinking (CT) based particle swarm optimization algorithm is put forward by introducing the CT capability, which belongs to human beings particularly. Based on the classical four stages theory for describing the CT process, the CTPSO algorithm framework and configuration is designed, and the evolution model is adapted. Focusing on the three-dimensional discrete decision variables and constrains in the model, the strategies of coding and decoding, constrains processing strategies and discrete strategy based on discrete space are designed. Experimental results show that CTPSO is superior in solution quality, convergence rate and robustness.(3) Distributed Weapon Target Assignment (DWTA) problem is described and analyzed, and an Extended Contract Net Protocol (ECNP) based problem solving framework is designed systemically. Announcement conditions, announcement strategy, bidding strategy, winner determination strategy and protocol are formalized and studied separately. The performing conditions of DWTA are given based on analysis of Remote Fire, and separation of targets performing authority. With the heavy traffic caused by announcement strategy of CNP, a receiver limited announcement strategy is put forward to reduce the set of the requirement receiver. Focusing on the local optimization of CNP, a bidding strategy based on replacement target and a protocol based on decommitment are put forward and combined to get higher quality of solutions. A winner determination strategy based on comprehensive evaluation is presented, with the measures of bidding value, global operation capability and communication cost. Simulation results show that with little more time cost, the ECNP based DWTA coordination mechanism can improve the effectiveness of NADMS and performing times of Remote Fire between weapons, meanwhile the communications in NADMS are reduced.(4) Distributed Guide Weapon Assignment (DGWA) problem is described and analyzed, as well as the performing conditions of it. According to the alterability of guide resources and the strict deadline of guidance coordination, a Conditional Contract Net Protocol (CCNP) based problem solving framework is designed systemically. The key techniques in auction and bidding mechanisms are pointed out. For the auction mechanism, based on analyzing the performing conditions, the corresponding coordination deadline and reserve prices models are designed. For the long coordination deadlines, a stages auction mechanism is presented to reduce the waste of guide resources with no challenge to the solution quality. For the bidding mechanism, according to the limitation and alterability of guide resources, the scheduling relationships between bidding tasks are analyzed, a maximum effectiveness first based scheduling model is designed, and a bidding mechanism based on adaptive scheduling is presented. Furthermore, constrains of bidding, models of calculating bidding value, and the process of performing bidding mechanism are studied. Simulations show that the CCNP based DGWA coordination mechanism is superior in NADMS effectiveness and viability, and effective to exert air defense missiles'potential powers.The dynamic IC formation method in NADMS is studied systematically in this dissertation. The distributed IC formation mechanisms are validated by using agent-based modeling and simulation techniques. We also analyze the operational superiority of NADMS that distributed IC formation mechanisms bring in. This research will support the development of NADMS in the future.